2 ms·
The hard statistical challenge is your voting theory is pretty much a proxy for urban/rural. So you'd need some sneaky statistics to find the rare areas where
by VLM 13y ago
The hard statistical challenge is your voting theory is pretty much a proxy for urban/rural. So you'd need some sneaky statistics to find the rare areas where you can separate the two closely matched data sources. Rephrased, I bet if you ran correlation coeff for urban/rural ratio vs both voting and death rates I bet they'd match up about the same and the coeff between urban/rural and death would be higher. Or I could be totally wrong and you're right. But what I am pretty sure I'm correct about, is this experiment would be the logical second step after your theory.